In 2010, I started a billion-dollar company while working 100 hours a week at my full-time job.

Here's how I'd do it in 2026 with just 5 hours and Claude AI. Back then, I thought my schedule was the problem.

It wasn't. I was using brute force where I needed systems.

Founders with day jobs think they need more time to build a business. That’s not it, they need leverage.

If I had to start over tomorrow with nothing but a laptop and Claude, here’s how I’d build a multi-million dollar business in 90 days.

I'd create a 5-person AI department that works 24/7.

Let me introduce you to the team… THE RESEARCHER

Upload a 50-75 page document explaining my vision and business model.

Ask:

"What problems are people complaining about in [industry] forums where existing solutions are failing them? Show me Reddit threads, Amazon reviews, and Google Trends data."

Follow Up:

"Find the top 5 urgent problems in growing industries with supporting evidence."

My team and I have been using Claude every day for months.

We've gotten it wrong more than we've gotten it right. But the mistakes taught us everything.

If you're switching to Claude right now (and a lot of people are), here's what we learned the hard way: MISTAKE #1: We used Claude like ChatGPT.

Short prompts. One-shot questions. Surface-level asks.

Claude isn't built for quick answers. Claude is built for deep thinking.

Instead of asking "what are the best marketing channels?" ask:

"Here's my business model, my target customer, and my current strategy. Where is my thinking weak? What am I not seeing?"

Then push back. "Too generic." "Be more specific to my market." "That's obvious. Go deeper."

How I'd 10x leads in 4 weeks using AI, if I were starting my business from scratch: (thread) Week 1:

Feed ChatGPT/Claude/Gemini my best post and say "Give me 20 versions of this hook that make people stop scrolling."

Goal: Find 3 posts that get 10x more engagement than anything I've written before.

I built and sold a company worth over a billion dollars.

If I lost it all tomorrow and had to rebuild from $0, I wouldn't apply for a single job.

I'd use AI + 1 hour a night for 90 days.

Here's the exact playbook:

🧵👇 Week 1: Make AI find your unfair advantage.

Everyone has 1-2 skills the market already pays for. Most people just can't see them clearly.

Feed your resume, past projects, and recommendations into ChatGPT.

Prompt: "Identify my 5 highest-value skills by market demand. For each: who pays for this, what problem it solves, and what a $2K-$5K offer looks like."

AI sees patterns in your career you've been too close to notice.

After selling my company for $1,000,000,000, early founders ask me the same question: "How do you know if an idea is worth pursuing?"

Here's the 5-step validation process that answers this: Most side hustlers spend 6 months building something nobody wants.

Waking up at 5am won't fix a broken offer. Notion, ClickUp, and Pomodoros won't validate demand.

Most side hustlers aren't underworking, they're building the wrong thing.

But productivity without clarity just speeds up failure.

Before you build a landing page, logo, or funnel, run this loop:

The 5-Step "Clarity Before Code" Validation Process

I cut my $300K marketing team down to one person.

Revenue went up 40%.

Here's the AI system that made this possible. An old-school marketing team is expensive and slow.

Not because the people aren't talented, they are.

But because every project requires constant back-and-forth.

Everyone wants to make money with content.
Almost nobody’s doing it right.

AI made it easier to publish, but that’s the trap!
Speed without taste creates what we call “AI Slop.”

If you’re serious about turning content into income, you need to build signal in a world drowning in noise.

Here’s what actually works: 1. Know what you stand for.

If your message isn’t clear, no amount of AI polish will save you.

AI can’t give you conviction. Only you can.

You have to know your philosophy, your lens on the world, and the specific person you’re trying to help.

This framework will expose every liar in your life.
Use it wisely.

Your business partner says "we value transparency" but hides the real numbers.
Your spouse says "communication matters" but shuts down every hard conversation.
Your team member says "I'm committed to growth" but won't take feedback.

Everyone has values. Few people live them.

And when you call them out, they get defensive. They make excuses. They tell you why this situation is different.

Here's what I do instead. I trap them in their own values.

Not to humiliate them. To force alignment between what they say and what they do.

But first, a warning: this is powerful. Use it to expose truth, not to win arguments or manipulate people into doing what you want..

This is Andrew Huberman.

He's a Stanford neuroscientist and a master of unlocking human potential.

One of the most brilliant people I’ve had the privilege to hang out with.

Here are 8 of my favorite moments from our chat 4 years ago: 1. Dopamine ≠ Pleasure

Dopamine isn’t the reward itself—it’s the neurochemical that MAKES you want to chase a goal.

The rat without dopamine still enjoys food but won't move to get it.

Andrew calls it the currency that’s moved our species forward, driving us to create & build:

The "secret" to building a $1B company isn't your product… it's who builds it with you.

Here are 6 interview questions I use to find A-players. The Skills Demonstration Question

"Here's a real challenge we're facing: [describe actual business challenge]. How would you approach solving this?"

Most candidates give theoretical answers.

A-players break down the problem instantly, identify constraints you missed, and provide structured approaches that show real expertise.

Look for strategic clarifying questions and relevant experience application.

If 1,000 people won't pay you $100, you don't have a business. You have an expensive hobby with a logo.

Here's the $100,000 validation test that will spare you months of wasted effort: Forget audience. Forget aesthetics.

Start with one brutal equation:
1,000 × $100 = $100,000

That's your test.

If you can't clearly answer:
- Who the 1,000 are
- Why they would pay
- What result they get
- How you'll reach them

You're not ready to build.
You're still mythologizing your idea instead of modeling it.

Here's the framework:

Your business can only scale to the level of your incompetence.

Here are the 8 leadership principles that separate CEOs who scale from those who plateau:

(thread) 1. Document Your Culture or Watch It Die

At Quest, I led by example when we had 50 people.
At 3,000 employees? Impossible.

The solution:
Write down exactly what excellence looks like.
Create a playbook that scales without you being in every room.
Your culture must survive your absence.

I used AI to reverse engineer how 12 billion-dollar companies got their first 1000 customers.

Here’s how to do it to find the opportunities in your market. (thread) Step 1: Find their origin story using AI

Most successful companies have completely rewritten their history. The polished story isn't the real story.

Use this prompt: "Find the earliest archived versions of [company name]'s website, press releases, and founder interviews from 2010-2015. What was their original positioning and first target market?"

AI can dig up old TechCrunch articles, Wayback Machine snapshots, and forgotten founder tweets that reveal the messy truth.

I use AI 365 days a year to run a multi-million dollar business.

These 5 rules will 10x the quality of your AI output:

(thread) Rule 1: Think autistic PhD student, not magic genie

AI is very smart but you need to be literal, detailed, and iterative.

Don't ask: "Help me with marketing"

Do ask: "Write 3 LinkedIn posts for entrepreneurs struggling with productivity. Each post needs a scroll-stopping first line, a brief story, and a soft CTA to download my time management guide."

Market research firms charge $30,000 for insights I get from ChatGPT in 30 minutes.

Here are the 7 prompts that turn AI into your $500/hour research analyst: Prompt 1: Blind Spot Analysis

"Analyze what [top 3 competitors] are NOT addressing in their marketing, product features, and customer communications. What problems do they ignore or downplay?"

This reveals opportunities they're leaving on the table.

If you have 60 minutes tonight, you can validate your business idea by tomorrow.

Here's the script: Most founders waste months building products nobody wants.

I used to do this too.

Built entire companies around assumptions. Lost millions on "great ideas" that failed in the market.

Then I learned to kill bad ideas in 60 minutes instead of 6 months.

Here's my exact validation script with the benchmarks that matter:

If 1,000 people won't pay you $100, you don't have a business. You have an expensive hobby with a logo.

Here's the $100,000 validation test that will spare you months of wasted effort: Forget audience. Forget aesthetics.

Start with one brutal equation:

1,000 × $100 = $100,000

That's your test.

I go to bed at 9 PM like it's a religion. This 'lazy' habit helps me work 92 hours a week.

The math:

- Bed at 9PM
- Up at 3AM naturally (no alarm)
- 18 productive hours M-F
- Plus a few “extra” hours on the weekends

Your body has an optimal sleep window. Most people never find it because they're fighting their biology with alarms and caffeine. I spent years forcing myself awake at "productive" times.
Now I let my body tell me when it's ready.
I outwork everyone while feeling completely rested.

How to Find Your Sleep Sweet Spot:

Everyone should treat AI like a brilliant, autistic, PhD student. Most are still using it like a broken search engine.

They ask one surface-level question and expect magic.
Then they get garbage output and think AI is trash.

If I sold my company today, I'd start my next multi-million dollar business in 90 days using only AI. Here's how:

I'd create 5 AI specialists that work 24/7 for a fraction of what a single employee costs.
Here's precisely how I'd structure my AI dream team: THE ANALYST

I'd upload a detailed vision document explaining my business model
I'd have it identify emerging market opportunities with hard data
I'd ask: "Find the top 3 underserved problems in profitable industries"
I'd demand specific research on customer behavior and spending patterns
I'd make it spot trends competitors are missing